Split off data package

[anseljh]

Hello! This is an interesting project. I think the most valuable part of it is going to end up being not the code, but rather the data that's collected about which categories licenses fall under, and which categories are deemed "compatible".

To that end, what do you think about splitting off the data as a separate package (e.g., a package of just JSON files)? That way it could be used more readily in other projects & languages, and draw a bigger contributor base beyond Rubyists.

I have seen this pattern a few times for repos that are really about data at their heart.

[anseljh]

Examples of a data package & a similar split (in progress):

• https://github.com/kemitchell/english-irregular-verbs
• Stop being API and website and just be dataset okfn/licenses#45

And a formal format for data packages:

• http://data.okfn.org/doc/data-package
• http://dataprotocols.org/data-packages/index.html

[andrew]

@anseljh that's a great idea, I've initially started by splitting the license arrays out into a separate JSON file: https://github.com/librariesio/license-compatibility/blob/master/lib/license/licenses.json

Some next steps:

• Extract the data into a separate repository
• Follow the data-package schema